Spark SQL data source

The data source

Spark SQL supports operating a variety of data sources through the SchemaRDD interface. A SchemaRDD can be operated as a general RDD or can be registered as a temporary table. R egistering a SchemaRDD for a table allows you to run SQL queries on its data. This section describes several ways to load data for SchemaRDD.

• RDDs
• Parquet file
• JSON dataset
• Hive table